Mutagonda et al. Malar J (2017) 16:267 DOI 10.1186/s12936-017-1914-9 Journal

RESEARCH Open Access Efect of pharmacogenetics on plasma and malaria treatment outcome in pregnant women Ritah F. Mutagonda1*, Appolinary A. R. Kamuhabwa1, Omary M. S. Minzi1, Siriel N. Massawe2, Muhammad Asghar3, Manijeh V. Homann3, Anna Färnert3,4 and Eleni Aklillu5

Abstract Background: Pregnancy has considerable efects on the pharmacokinetic properties of drugs used to treat uncom- plicated falciparum malaria. The role of pharmacogenetic variation on anti-malarial drug disposition and efcacy during pregnancy is not well investigated. The study aimed to examine the efect of pharmacogenetics on lumefantrine (LF) pharmacokinetics and treatment outcome in pregnant women. Methods: Pregnant women with uncomplicated falciparum malaria were enrolled and treated with - lumefantrine (ALu) at Mkuranga and Kisarawe district hospitals in Coast Region of Tanzania. Day-7 LF plasma con- centration and genotyping forCYP2B6 (c.516G>T, c.983T>C), CYP3A4*1B, CYP3A5 (*3, *6, *7) and ABCB1 c.4036A4G were determined. Blood smear for parasite quantifcation by microscopy, and dried blood spot for parasite screening and genotyping using qPCR and nested PCR were collected at enrolment up to day 28 to diferentiate between reinfec- tion from recrudescence. Treatment response was recorded following the WHO protocol. Results: In total, 92 pregnant women in their second and third trimester were included in the study and 424 samples were screened for presence of P. falciparum. Parasites were detected during the follow up period in 11 (12%) women between day 7 and 28 after treatment and PCR genotyping confrmed recrudescent infection in 7 (63.3%) women. The remaining four (36.4%) pregnant women had reinfection: one on day 14 and three on day 28. The overall PCR- corrected treatment failure rate was 9.0% (95% CI 4.4–17.4). Day 7 LF concentration was not signifcantly infuenced by CYP2B6, CYP3A4*1B and ABCB1 c.4036A>G genotypes. Signifcant associations between CYP3A5 genotype and day 7 plasma LF concentrations was found, being higher in carriers of CYP3A5 defective variant alleles than CYP3A5*1/*1 genotype. No signifcant infuence of CYP2B6, CYP3A5 and ABCB1 c.4036A>Genotypes on malaria treatment outcome were observed. However, CYP3A4*1B did afect malaria treatment outcome in pregnant women followed up for 28 days (P 0.018). = Conclusions: Genetic variations in CYP3A4 and CYP3A5may infuence LF pharmacokinetics and treatment outcome in pregnant women. Keywords: Malaria in pregnancy, Day 7 lumefantrine concentration, Recrudescence, Pharmacogenetics

Background drugs, usually resulting in lower drug exposures and Pregnancy-induced physiological changes alter the phar- lower cure rates especially in advanced pregnancy stage macokinetic properties of a number of anti-malarial compared to non-pregnant population [1, 2]. Preg- nancy-associated physiological changes result in lower *Correspondence: [email protected] drug absorption, enhances drug clearance, and increase 1 Department of Clinical Pharmacy and Pharmacology, School body fuid volume of distribution [3–5]. Drug exposure of Pharmacy, Muhimbili University of Health and Allied Sciences, depends greatly on the rate of metabolism and diferences P.O. Box 65013, Dar es Salaam, Tanzania Full list of author information is available at the end of the article in activity of metabolizing enzymes can signifcantly

© The Author(s) 2017. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Mutagonda et al. Malar J (2017) 16:267 Page 2 of 10

alter the efcacy of drugs. Elevated levels of oestrogens, treated with ALu in Tanzania. Te fndings might have , cortisol, and prolactin during pregnancy implications for treatment policies of not only anti- have been linked to altered expression and metabolic malarial drugs, in particular the widely used artemisinin- activity of several hepatic enzymes. based combinations, but also other drugs metabolized by For instance, catalytic activity of CYP3A4, CYP2C9 and these enzymes. CYP2A6 enzymes increases during pregnancy, while CYP2C19 and CYP1A2 enzyme activity decreases [6, Methods 7]. Tese enzymes are involved in metabolism of several Study design and population anti-malarial drugs including LF and artemether [8–10], Tis was a one-arm prospective cohort study that and are genetically polymorphic displaying wide inter- included all pregnant women who gave consent to par- individual variations in enzyme activity [11–16]. ticipate in the study when attending antenatal clinics at Lower drug exposure levels in pregnant women have Kisarawe and Mkuranga district hospitals, northern Tan- been reported for artemether/, zania. Between May 2014 to April 2015, pregnant women /dihydroartemisinin, dihydroartemisinin and attending the antenatal clinics (ANCs) were screened lumefantrine (LF) [1, 17–19]. Tis increases the risks for malaria infection by using malaria rapid diagnos- of treatment failure, increase risk of adverse outcomes tic test (MRDT). Pregnant women with uncomplicated for the fetus associated with malaria complications, and infection and haemoglobin development of resistance to malaria parasites. A higher level of >8 g/d were enrolled. Te study received ethics treatment failure rate has indeed been observed in preg- approval from the institutional review board of Muhim- nant women compared to non-pregnant ones living in bili University of Health and Allied Sciences (MUHAS). the same area [1, 19, 20]. In this case, treatment failure Participants were informed about the aim of the study may not be caused by intrinsic parasite resistance but is and gave written consent before participating in the instead the result of inadequate drug levels due to preg- study. To ensure confdentiality, women’s identifcation nancy, pharmacogenetic profle of the host or other non- numbers were used when labelling samples and during genetic modifers of the pharmacokinetic parameters. data entry into confdential report forms (CRF). Genetic variation in drug metabolizing enzymes and transporter proteins might predict plasma exposure Sample size and treatment failure and/or emergence of drug resist- Considering anticipated population proportion (P) of ant pathogens on infectious and non-infectious diseases, clinical failures in pregnant women being 18% [19], with such as malaria, HIV and tuberculosis [20–24]. CYP3A 95% confdence level and 10% precision, 92 malaria posi- responsible for metabolism of artemether and LF is tive pregnant women were enrolled in this study. induced by approximately twofold during the third tri- mester of human pregnancy. Low cure rates (83.5%) have Treatment, clinical procedures and follow‑up been reported for pregnant women in Tailand receiving Te study participants received six doses of four tablets ALu [2]. However, pregnant women in Uganda showed of ALu ­(Coartem®; Pharma AG, Basel, Switzer- an adequate clinical response (98.2%) using similar doses land) (20 mg artemether and 120 mg lumefantrine) over of ALu [25]. Te diferences in these studies might be the course of 3 days at 0, 8, 24, 36, 48, and 60 h. For each explained by diferences in host genetics, pharmacokinet- patient, general physical examination was performed at ics, diferent resistance patterns of malaria parasites, or enrollment (day 0) and on follow up visits on days 2, 7, higher levels of background immunity among individual 14, 21 and 28. Approximately 50 μl of blood was collected pregnant women [25]. Another study reported that the on flter paper (Whatman grade 3) for later PCR analysis. wide diference in CYP3A4*1B allele frequency between Each flter paper was dried and individually stored in a the Tanzania and populations presents a plastic bag and kept frozen at −80 °C. At enrollment day, potential explanation for the lower efcacy of ALu in 1 ml of whole blood was taken into an EDTA containing Cambodia and highlighted the importance of pharmaco- vacutainer tube and stored at −80 °C at MUHAS labora- genetic considerations in the decision-making process of tory until further analysis. Additionally, 3 mls of venous frst-line treatment policies for specifc populations [8]. blood were drawn from pregnant women in heparinized To date, there are limited studies that attempted to tubes on day 7 to determine plasma LF concentrations. examine the role of pharmacogenetics on the pharma- cokinetics of anti-malarial drugs and treatment outcomes CYP3A4, CYP3A5, CYP2B6, and ABCB1 genotyping in pregnant women. Te aim of this study is to investigate Genomic DNA was isolated from peripheral leuko- the efects of pharmacogenetics on day 7 LF plasma con- cytes in whole blood samples using QIAamp DNA Midi centrations and treatment outcome in pregnant patients Kit (Qiagen GmbH, Hilden, Germany) according to the Mutagonda et al. Malar J (2017) 16:267 Page 3 of 10

manufacturer’s instructions. Genotyping for the common PCR genotyping to diferentiate recurrent P. falcipa- functional variant alleles for CYP2B6*6, CYP2B6*18, rum infections from reinfections were done according to CYP3A4*1B, CYP3A5*3, CYP3A5*6, CYP3A5*7 and World Health Organization (WHO) recommendations ABCB1 c.4036A>G (rs3842), which have been reported [29], by characterizing the length polymorphism of the to be relevant for artemether and LF disposition [26, merozoite surface protein 2 gene (msp2) in samples col- 27] were done as described previously [12, 22]. In brief lected at day 0 and on the day recurrent parasitaemia was genotyping was performed using TaqMan drug metabo- found. Recrudescence was determined when at least one lism genotyping assay reagents for allelic discrimination msp2 allele of the same allelic type and of identical base (Applied Biosystems Genotyping Assays) with the fol- pair size was found in samples collected on day 0 and on lowing ID numbers for each SNP: C__7817765_60 for the day of recurrent infection. A reinfection was defned CYP2B6*6 (c.516G4T, rs3745274), C__60732328_20 for as all alleles were of diferent length in the sample col- CYP2B6*18 (c.983T4C, rs28399499), C__26201809_30 lected on day 0 and that collected on the day of recurrent for CYP3A5*3 (c.6986A4G, rs776746), C__30203950_10 infection. for CYP3A5*6 (g.14690G4A,rs10264272) and C__32287188_10 for CYP3A5*7 (g.27131_27132insT Quantifcation of LF plasma concentrations rs41303343) and C__11711730_20) for ABCB1 Blood samples for the determination of LF plasma con- c.4036A>G (rs3842), and C__11711730_20 for centrations were collected on day 7 (corresponding to CYP3A4*1B (−392A>G, rs2740574). Genotyping was 168 h) following initiation of ALu treatment. Blood sam- carried out using Quant Studio 12 K Flex Real-Time PCR ples for quantifcation of LF levels were centrifuged and system (Life Technologies Holding, Singapore, Singa- plasma was stored at −80 °C until analysis. Plasma LF pore). Te fnal volume for each reaction was 10 μl, con- concentration was analysed using a validated method of sisting of TaqMan fast advanced master mix (Applied high performance liquid chromatography (HPLC) with Biosystems, Waltham, MA, USA), TaqMan 20X drug ultraviolet detection at Sida/MUHAS bioanalytical labo- metabolism genotyping assays mix (Applied Biosystems) ratory in Dar-es-Salaam, Tanzania [30]. Te coefcients and genomic DNA. Te PCR profle consisted of an ini- of variation (CV %) during the analysis of LF were 8.4, 4.7 tial step at 60 °C for 30 s, hold stage at 95 °C for 10 min and 4.5% at 100, 1000, and 8000 ng/ml, respectively. Te and PCR stage for 40 cycles step 1 with 95 °C for 15 and lower limit of quantifcation was 50 ng/ml. step 2 with 60 °C for 1 min and after read stage with 60 °C for 30 s. Data analysis LF plasma concentration data were log transformed to Parasite detection and genotyping achieve normality of data distribution. Median (inter- Dried blood spots on flter papers obtained at enrolment quartile range) was used to describe day-7 LF plasma (day 0) and on follow-up days (day 2, 7, 14, 21 and 28) concentrations. Comparison of day-7 median LF plasma were punched, and one circle 5 mm in diameter was used concentrations in pregnant women between the diferent for DNA extraction using QIAamp DNA blood micro kit genotypes were carried out using Kruskal–Wallis one- (Qiagen GmbH, Hilden, Germany) following the manu- way analysis of variance (ANOVA). X­ 2 test was used to facturer’s recommendations. Detection of Plasmodium compare the observed and expected allele frequencies parasites was performed using a species-specifc PCR according to the Hardy–Weinberg equilibrium. Infuence targeting the ssRNA gene [28]. PCRs were carried out of human genotype on malaria treatment outcome was in duplicate in a 25-μl fnal volume containing 12.5 μl of analysed using Pearson’s Chi square and Fisher’s exact Universal PCR Master Mix, 5 μl of DNA, forward and test. Haploview software package (version 4.2) was used reverse primers at various concentrations, and P. falcipa- to analyse linkage disequilibrium (LD) and haplotype rum, , and Plas- construction. Statistical analyses were performed using modium ovale probes at a fnal concentration of 100 nM. Statistical Package for Social Sciences (SPSS) software, All reactions were run on an ABI Prism 7000 sequence version 22.0 (IBM Corporation, Somers, NY, USA). P val- detection system (Applied Biosystems) with the default ues <0.05 were considered to be statistically signifcant. settings; each sample was initially denatured at 95 °C for Malaria treatment outcomes were classifed following 10 min and cycled 40 times, with each cycle consisting the WHO protocol [31], as adequate clinical and para- of 95 °C for 15 s and 60 °C for 60 s. Each reaction plate sitological response (ACPR), corrected for reinfection included four positive controls for P. falciparum, P. ovale, using PCR genotyping on day 28 and treatment failure P. vivax, and P. malariae and a negative control with (TF); designated as early treatment failure (ETF), late molecular-grade water in place of DNA. clinical failure (LCF), or late parasitological failure (LPF). Mutagonda et al. Malar J (2017) 16:267 Page 4 of 10

Te infuence of genetic variations on malaria treatment PCR-corrected ACPR rate as defned by absence of para- outcome was evaluated using Cox regression analysis. sitaemia on day 28, irrespective of axillary temperature, in patients who did not previously meet any of the crite- Results ria of ETF, LCF or LPF was 91.0% (95% CI 82.62–95.58). Patient characteristics Te rate of PCR-corrected LPF, as defned by presence of In total, 1835 pregnant women were screened using parasitaemia on any day between day 7 and 28 in patients MRDT and a total of 92 pregnant women with malaria who did not previously meet any of the criteria of ETF or infection consented and were enrolled in the study. Base- LPF, was 9.0% (95% CI 4.42–17.38). Tere was no ETF or line characteristics are presented in Table 1. Te median LCF. age of pregnant patients was 23 (range 15–41) years and the majority (54.3%) were multigravida. Most partici- Allele, genotype and haplotype frequency distributions pants were in the second trimester (60.7%). Te median Te overall CYP2B6*6, CYP2B6*18, CYP3A4*1B, parasite density was 2700 (range 400–72,500) parasites/ CYP3A5*3, CYP3A5*6, CYP3A5*7, ABCB1 c.4036A>G µl. genotype and allele frequencies in Tanzanian pregnant women is presented in Table 2. Tere was no signifcant Day 28 malaria treatment outcome in pregnant women deviation between the observed and expected genotype In total, 424 samples were screened for presence of para- frequencies from Hardy–Weinberg equilibrium. Te var- sites and P. falciparum was present in 11 (12%) patients iant allele frequency was highest (76.1%) for CYP3A4*1B during the follow up period. Two samples were detected followed by CYP2B6 c.516G>T (*6, 33.5%). CYP2B6 on day 7, four on day 14, one on day 21 and four on day 28 c.983T>C (*18) had the lowest allele frequency which after ALu treatment. PCR based genotyping of msp2 con- was 9.3%. Genotype of CYP450 enzymes and ABCB1 frmed recrudescent infection in seven women (63.3%), transporter was equally distributed among the age, grav- two on day 7, three on day 14, one on day 21 and one on ida, trimester and PMTCT status of pregnant women day 28. Te remaining four (36.4%) pregnant women had (P > 0.05). reinfection; one on day 14 and three on day 28. Te over- Haplotype analysis of CYP3A4*1B (–392A>G), all rate of reinfection was 4.3% (95% CI 0.15–8.45). CYP3A5*3 (g.6986A>G), CYP3A5*6 (g.14690G>A), PCR uncorrected ACPR on day 28 was 88.9% (95% CI CYP3A5*7 (27131_27132insT) is presented in Fig. 1 and 82.06–95.74) and TF rates were 11.1% (95% CI 4.26– Table 3. Tere was no linkage between the two CYP2B6 17.94). Rate of ACPR was calculated using per-protocol variant alleles (c.516G>T and c.983T>C). Likewise, no LD method, where patients were excluded due to lost to between the three CYP3A5 SNPs (CYP3A5*3, CYP3A5*6, follow-up, protocol violations and TF due to reinfection. CYP3A5*7) was found. Instead each SNP was inversely linked to each other (i.e., each SNP is in strong LD with the wild type variant of the other two SNPs). Te haplo- Table 1 Baseline characteristics of study participants type frequency of CYP3A5*1, *3, *6 and *7 was 44.5, 22.6, (n 92) 20.7 and 12.2%, respectively. Interestingly all the three = Characteristic Number of pregnant Percentage CYP3A5 variant alleles (CYP3A5*3, CYP3A5*6 and *7) women occur in high LD with CYP3A4*1B. Te major CYP3A haplotype was CYP3A4*1B alone (34.2%) followed by its Age (years) linkage with CYP3A5*6 (17.6%) and CYP3A5*3 (13.3%) <18 8 8.7 (Table 3). 19–25 52 56.5 >25 32 34.8 The efect of genotype on day 7 LF concentration Gravida (n) in pregnant women Primigravida 42 45.7 Based on the haplotype analysis result, subjects were Secundigravida 17 18.5 grouped according to the numbers of functional Multigravida 33 35.9 CYP2B6*1 and CYP3A5*1 variant alleles; two (*1/*1), Trimester (n) one (heterozygous for defective variant alleles) and zero Second 66 60.7 (homozygous for defective variant alleles) to investigate Third 26 39.3 the efect of genotype on day 7 plasma LF concentrations. Parasitaemia (parasites/µl blood) Te median day 7 plasma LF concentration was 650 <1000 14 15.5 with the range of 2936 (124–3059) ng/ml. Comparison 1000–10,000 51 55.0 of the median plasma LF concentrations between the >10,000 27 29.5 diferent genotypes is presented in Table 4. Tere was Mutagonda et al. Malar J (2017) 16:267 Page 5 of 10

Table 2 Genotype and variant allele frequency distribu- tion among pregnant women with uncomplicated Plasmo- dium falciparum infection in Tanzania

Genotype Frequency N (%)

CYP2B6 c.516G>T (*6) *1/*1 38 (46.3%) *1/*6 33 (40.2%) *6/*6 11 (13.4%) CYP2B6 c.983T>C (*18) *1/*1 75 (82.4%) *1/*18 15 (16.5%) *18/*18 1 (1.1%) CYP3A4*1B ( 392A>G) − *1/*1 3 (3.2%) *1/*1B 38 (41.3%) *1B/*1B 51 (55.4%) CYP3A5*3 c.6986A>G *1/*1 51 (56.6%) *1/*3 37 (41.1%) *3/*3 2 (2.2%) CYP3A5*6 c.14690G>A *1/*1 59 (64.1%) *1/*6 28 (30.4%) *6/*6 5 (5.4%) CYP3A5*7 27131_27132insT *1/*1 68 (75.6%) *1/*7 22 (24.4%) *7/*7 0 ABCB1 c.4036AG (rs3842) Fig. 1 Linkage disequilibrium (LD) plot of CYP3A4*1B ( 392A>G), AA 54 (58.7%) − CYP3A5*3 (g.6986A>G), CYP3A5*6 (g.14690G>A), CYP3A5*7 AG 26 (28.3%) (27131_27132insT) and the observed D′ values (within the diagonal GG 12 (13.4%) boxes). The pair-wise LD relationship between two SNPs and the Allele Minor allele % respective D′ value is indicated in each square. The color gradient from red to white reveals higher to lower LD (D′ 1–0) values CYP2B6 c.516G>T (*6) *6 33.5 CYP2B6 c.983T>C (*18) *18 9.3 CYP3A4*1B *1B 76.1 CYP3A5*3 *3 22.8 having plasma LF concentration ≥600 ng/ml (Pearson CYP3A5*6 *6 20.6 Chi square test, P = 0.01). Te proportion of subjects CYP3A5*7 *7 12.2 with two functional CYP3A5 genotype (*1/*1) was signif- ABCB1 c.4036AG (rs3842) G 27.2 cantly higher among those with plasma LF conc <600 ng/ ml (34.6%) compared to those with ≥600 ng/ml (5.9%). On the other hand proportion of subjects with lack- ing functional CYP3A5 (homozygous for defective no signifcant efect of CYP2B6, CYP3A4*1B or ABCB1 variant alleles) was higher among those with plasma c.4036AG genotype on day 7 LF plasma concentrations. LF conc ≥600 ng/ml (38.2%) compared to those with Te median plasma LF concentration was signifcantly >600 ng/ml (19.2%). associated with CYP3A5 genotypes (P = 0.039). CYP3A5 defective allele carriers displayed a higher plasma LF The efect of genotype on malaria treatment outcome concentrations compared to those homozygous for in pregnant women The possible influence of genetic variations in CYP3A5*1/*1 genotypes (P = 0.04). Tere was signif- cant association between having CYP3A5 genotype and CYP2B6, CYP3A5 and ABCB1 on malaria treatment Mutagonda et al. Malar J (2017) 16:267 Page 6 of 10

Table 3 CYP3A haplotype distribution among pregnant women with uncomplicated Plasmodium falciparum infection in Tanzania CYP3Ahaplotype CYP3A SNP combinations Frequency (%) CYP3A4*1B CYP3A5 *3 CYP3A5 *6 CYP3A5 *7 392A>G g.6986A>G g.14690G>A 27131_27132insT − GAGA *1B wt Wt wt 34.2 GAAA *1B wt *6 wt 17.6 GGA 1B *3 Wt wt 13.3 AAGA wt wt Wt wt 11.4 GAGT *1B wt Wt *7 11.0 AGGA wt *3 Wt wt 8.2 AAAA wt wt *6 wt 3.0 AGGT wt wt Wt *7 1.3 wt wild type variant allele, SNP single-nucleotide polymorphism

Table 4 Comparison of median and inter quartile range (IQR) Discussion of day 7 LF median plasma concentration between diferent In the present study, the efect of pharmacogenetics genotype groups in pregnant women with uncomplicated on LF pharmacokinetics (focusing on day 7 LF plasma Plasmodium falciparum infection using Kruskal–Wallis ANOVA concentration, which is a surrogate marker for AUC Plasma lumefantrine concertation at day 7 and malaria cure rate) and treatment outcome in preg- n Median (IQR) P value nant women was investigated. For this purpose, func- tional variant alleles in CYP450 (CYP2B6, CYP3A4, No of CYP2B6*1 allele CYP3A5) and ABCB1 genes involved in the metabolism Zero 17 650.0 (285.8–1603.2) 0.85 of anti-malarial drugs were assessed in pregnant malaria One 37 671.6 (280.0–1935.3) patients treated with ALu. Te major fnding includes: Two 28 672.0 (350.0–2161.8) (i) CYP3A5 genotype has signifcant infuence on plasma CYP3A4*1B LF concentration and (ii) CYP3A4*1B is associated with *1A/*1A 3 294.6 (270.0–319.2) 0.24 malaria treatment outcome. A number of studies have *1A/*1B 38 689.7 (354.1–1935.3) speculated that lower artemether and LF concentration *1B/*1B 51 644.6 (280.0–2615.0) in pregnant women has been due to changes in catalytic No of CYP3A5*1 allele activities of CYP450 metabolizing enzymes particularly Zero 27 716.6 (340.9–2109.4) 0.039 the increased activity of CYP3A4 [19, 20, 32]. Tis is the One 47 689.7 (280.0–3059.1) frst study in pregnant women to investigate the efect Two 18 354.2 (285.8–1603.2) of pharmacogenetics on day 7 LF concentrations and ABCB1 c.4036AG malaria treatment outcome after a follow up of 28 days. AA 54 585.2 (330.0–2615.0) 0.28 Te observed variant allele frequencies for CYP2B6*6 AG 26 710.7 (282.3–2161.8) (c.516G→T), CYP3A4*1B, and CYP3A5*3 are very sim- GG 12 1055.1 (285.8–2109.4) ilar to those previously reported from previous studies No of CYP2B6*1 allele: two, *1/*1; one, heterozygous for *6 or *18; zero, conducted in Tanzanian population [8, 12, 22, 23]. homozygous for *6 or *18 or combination thereof Lumefantrine, the long-acting component of the most No of CYP3A5*1 allele: two, *1/*1; one, heterozygous for *3, *6 or *7; zero, homozygous for *3, *6 or *7, combination thereof widely used artemisinin-based (ACT) in Africa is metabolized by CYP3A4. Artemether is metabolized by CYP3A4 and CYP3A5. Terefore, outcome was evaluated using Cox regression analy- information on the role of pharmacogenetics on drug sis. A log rank test was run to determine differences disposition and treatment outcome in pregnant women in the cumulative hazard distribution between the dif- using ALu is needed. CYP3A5 is mainly expressed in ferent genotypes. There were no associations between black population and its genotype contributes to vari- malaria treatment outcome and ABCB1 transporter ations in the total CYP3A enzyme activity as measured and most of CYP450 genotypes with an exception to by 4beta-hydroxycholesterol, an endogenous CYP3A CYP3A4 (Table 5). marker [13, 15, 33, 34]. Mutagonda et al. Malar J (2017) 16:267 Page 7 of 10

Table 5 Analysis of genetics efect on malaria treatment outcome (risk of recrudescence) in pregnant women Genotype No of patients No. of recrudescence Percentage (%) P value (log-rank test)

CYP2B6*6 0.676 CYP2B6*1/*1 35 1 2.8 CYP2B6*1/*6 32 2 6.5 CYP2B6*6/*6 11 1 9.0 CYP2B6*18 0.605 CYP2B6*1/*1 72 5 6.9 CYP2B6*1/*18 14 2 14.2 CYP2B6*18/*18 1 0 0 CYP3A4*1B 0.018 CYP3A4*1/*1 2 1 50 CYP3A4*1/*1B 37 3 8.1 CYP3A4*1B/*1B 49 3 6.1 CYP3A5*3 0.087 CYP3A5*1/*1 48 3 6.2 CYP3A5*1/*3 36 3 8.3 CYP3A5*3/*3 2 1 50 CYP3A5*6 0.645 CYP3A5*1/*1 57 4 7.0 CYP3A5*1/*6 26 3 11.5 CYP3A5*6/*6 5 0 0 CYP3A5*7 0.134 CYP3A5*1/*1 66 7 10.6 CYP3A5*1/*7 20 0 0 CYP3A5*7/*7 – – – ABCB1 rs3842 0.462 A/A 12 0 0 A/G 25 3 12 G/G 51 4 7.8

Tere was signifcantly higher plasma LF concentration caused by a decrease of 36.5% in the inter-compartmental in patients carrying CYP3A5 defective alleles than those clearance in pregnant women [18]. Te decrease in clear- without. Tough not signifcant, a similar fnding was ance can be attributed to changes in body composition observed in a recent HIV-malaria cohort study in Tan- (increased plasma volume, water, and fat content) during zania [22]. Pregnant women withCYP3A5*1/*1 genotype pregnancy [36] where the distribution of the lipophilic had signifcantly higher risk of having LF plasma concen- LF through the body might be altered substantially. In tration <600 ng/ml. Plasma LF concentration <600 ng/ another study conducted in Tanzania it was also reported ml is associated with risk of recurrent parasitaemia in that CYP3A4*1B genotype did not afect day 7 LF plasma pregnant women [20]. However, there was no signif- levels unless it was induced by a potent CYP3A4 inducer cant impact of CYP3A5 genotype on malaria treatment [22]. outcome which is similar to previous reports [22]. Tis Te study also analysed patients with recrudescence may indicate that CYP3A5 plays a minor role compared and compared their pharmacogenetic profles with to CYP3A4 in determining malaria treatment outcome. those with an ACPR. Tere was a signifcant asso- Indeed, CYP3A4 is the major drug metabolizing enzyme ciation between CYP3A4*1B genotype and treatment than CYP3A5 for most CYP3A substrate drugs [35]. outcome in pregnant women. It was observed that In this study, there was no infuence of CYP3A4*1B majority of pregnant women with CYP3A4*1B/*1B had on day 7 LF plasma levels. Te fndings are similar to attained ACPR compared to those with CYP3A4*1/*1. previous reports from Cambodia and Tanzania [8]. A Inconclusive data on alterations of enzyme activ- study conducted in Uganda reported that lower LF day 7 ity in CYP3A4*1B carriers are reported in the litera- plasma concentrations observed during pregnancy were ture. Some investigators have suggested CYP3A4*1B Mutagonda et al. Malar J (2017) 16:267 Page 8 of 10

is associated with increased CYP3A4 expression and in the CYP3A locus for LF pharmacokinetics and treat- enhanced drug elimination in carriers of CYP3A4*1B ment outcome in pregnant women. Tis study also pre- may lead to treatment failure [37]. In contrast, the sents haplotype structure of the most common CYP3A association of CYP3A4*1B with lower CYP3A4 enzyme functional variant alleles in African population for future activity in Tanzanians is reported previously, where genetic association studies. carriers of CYP3A4*1B had a signifcantly lower enzyme activity than CYP3A4*1 [38]. Te fnding of Conclusions signifcantly lower total CYP3A activity in Tanzanians In general, the study fnding indicates association of than whites (Swedes) and Asians (Koreans) despite hav- the low enzyme activity genotype of CYP3A4*5 and ing high allele frequency ofCYP3A4*1B in Tanzanians CYP3A4*1B with high LF plasma exposure and bet- (77%) [34] may indicate the association of CYP3A4*1B ter malaria treatment outcome, respectively. Te efects with low enzyme activity in blacks. CYP3A4*1B variant of pharmacogenetics on LF pharmacokinetics were well allele is absent in Asians and occurs at a much lower characterized in pregnant patients with uncomplicated P. frequency (2–9%) in whites [39]. falciparum malaria. Te importance of these fndings is Preliminary fnding indicates that pharmacogenetic that in the future genotyping can be used to predict the variation in CYP3A4 and CYP3A5 infuences the LF need for anti-malarial drugs dosage adjustment to preg- plasma exposure and malaria treatment outcome in nant women with CYP3A4*1/*1or CYP3A5*1/*1. Te pregnant women. Since CYP3A is responsible for the impact of CYP3A haplotypes on the metabolism of anti- metabolism of artemether and LF, interplay between malarial drugs in a large sample size cohort needs to be CYP3A4 and CYP3A5 genotypes may determine ACT further evaluated. plasma exposure and treatment outcome. CYP3A4 and Authors’ contributions CYP3A5 haplotypes are located in the same gene locus, AARK EA and SSM conceived the study, participated in the study protocol efects initially considered to be due to a CYP3A4 allele development, coordination and manuscript writing. OMSM, AF, AM, EA and MVH participated in the data analysis as well as manuscript writing. RFM might actually be due to a CYP3A5 allele in LD [40]. participated in study protocol development, data collection and analysis and LD between CYP3A4*1B and CYP3A5*1Ais suggested in the preparation and writing of the manuscript. All authors participated in as possible cause of inter individual variation in CYP3A reading. All authors read and approved the fnal manuscript. metabolism [41, 42]. Author details Previous studies in white population reported that 1 Department of Clinical Pharmacy and Pharmacology, School of Pharmacy, CYP3A4*1B is linked with the functional CYP3A5*1 Muhimbili University of Health and Allied Sciences, P.O. Box 65013, Dar es Salaam, Tanzania. 2 Department of Obstetrics and Gynaecology, School resulting in high enzyme activity [42, 43]. In contrast of Medicine, Muhimbili University of Allied Sciences, P.O Box 65013, Dar es our extensive haplotype analysis in Tanzanians (Fig. 1; Salaam, Tanzania. 3 Unit of Infectious Diseases, Department of Medicine, Table 3) indicates that CYP3A4*1B is linked with CYP3A5 Karolinska Institutet, Solna, 171 76 Stockholm, Sweden. 4 Department of Infec- tious Diseases, Karolinska University Hospital, Stockholm, Sweden. 5 Section defective variant alleles (*3, *5 and *7) and hence may of Pharmacogenetics, Department of Physiology and Pharmacology, Karolin- be associated with low enzyme activity. It is well known ska Institutet, 171 77 Stockholm, Sweden. that sub-Saharan African population is the most geneti- Acknowledgements cally heterogeneous population globally, characterized by We are grateful to our fellow malaria team members at MUHAS-Dr. Billy extensive population substructure and unique LD pattern Ngasala, Dr. ElifordNgaimisi and Dr. Edna Gasarasi for providing assistance. We compared to non-African populations [44, 45]. Lack of thank Dorisia Nanage for her dedicated technical assistance. We also thank the health care providers and the study participants from Mkuranga and Kisarawe LD between the two CYP2B6 variant alleles (c.516G>T district hospitals for all the co-operation they gave to us throughout the study and c.983T>C) and between the three CYP3A5 SNPs period. (c.6986A4G, g.14690G4A and g.27131_27132 insT) in this Competing interests study is similar to previously reports from Africa [11, 13, The authors declare that they have no competing interests. 15, 23]. Limitation of this study includes that the sample size Availability of data and materials The datasets used and analysed during the current study is available from the was calculated based on the previous study whereby corresponding author on reasonable request. treatment failure was reported to be 18% which is higher than the failure rate reported in this study. As a Ethics approval and consent to participate The study received ethics approval from the institutional review board of result, although the sample size is enough to investigate Muhimbili University of Health and Allied Sciences (MUHAS). Participants were anti-malarial drug efcacy in the population, the study informed about the aim of the study and gave written consent before partici- is under-powered to investigate the impact of CYP3A pating in the study. To ensure confdentiality, women`s identifcation numbers were used when labeling samples and during data entry into confdential haplotypes on treatment outcome. However, the study report forms (CRF). fnding highlights the importance of genetic variation Mutagonda et al. Malar J (2017) 16:267 Page 9 of 10

Funding 17. Tarning J, Rijken MJ, McGready R, Phyo AP, Hanpithakpong W, Day NP, This study was supported by grants from the Swedish International Develop- et al. Population pharmacokinetics of dihydroartemisinin and pipe- ment Cooperation Agency Programme (SIDA) under MUHAS Reproductive raquine in pregnant and nonpregnant women with uncomplicated health projects. malaria. Antimicrob Agents Chemother. 2012;56:1997–2007. 18. Kloprogge F, Piola P, Dhorda M, Muwanga S, Turyakira E, Apinan S, et al. Population pharmacokinetics of lumefantrine in pregnant and nonpreg- Publisher’s Note nant women with uncomplicated Plasmodium falciparum malaria in Springer Nature remains neutral with regard to jurisdictional claims in pub- Uganda. CPT Pharmacometr Syst Pharmacol. 2013;2:e83. lished maps and institutional afliations. 19. Mosha D, Guidi M, Mwingira F, Abdulla S, Mercier T, Decosterd LA, et al. Population pharmacokinetics and clinical response for artemether-lume- Received: 10 January 2017 Accepted: 27 June 2017 fantrine in pregnant and nonpregnant women with uncomplicated Plas- modium falciparum malaria in Tanzania. Antimicrob Agents Chemother. 2014;58:4583–92. 20. Mutagonda RF, Kamuhabwa AA, Minzi OM, Massawe SN, Maganda BA, Aklillu E. Malaria prevalence, severity and treatment outcome in relation References to day 7 lumefantrine plasma concentration in pregnant women. Malar J. 1. McGready R, Stepniewska K, Lindegardh N, Ashley EA, La Y, Singhasi- 2016;15:278. vanon P, et al. The pharmacokinetics of artemether and lumefantrine 21. Maganda BA, Ngaimisi E, Kamuhabwa AA, Aklillu E, Minzi OM. The infu- in pregnant women with uncomplicated falciparum malaria. Eur J Clin ence of and -based anti-retroviral therapy on the Pharmacol. 2006;62:1021–31. pharmacokinetics of lumefantrine and anti-malarial dose recommenda- 2. McGready R, Tan SO, Ashley EA, Pimanpanarak M, Viladpai-Nguen J, Phai- tion in HIV-malaria co-treatment. Malar J. 2015;14:179. phun L, et al. A randomised controlled trial of artemether-lumefantrine 22. Maganda BA, Minzi OM, Ngaimisi E, Kamuhabwa AA, Aklillu E. CYP2B6*6 versus artesunate for uncomplicated Plasmodium falciparum treatment in genotype and high efavirenz plasma concentration but not nevirapine pregnancy. PLoS Med. 2008;5:e253. are associated with low lumefantrine plasma exposure and poor treat- 3. Pavek P, Ceckova M, Staud F. Variation of drug kinetics in pregnancy. Curr ment response in HIV-malaria-coinfected patients. Pharmacogenom J. Drug Metab. 2009;10:520–9. 2016;16:88–95. 4. Ward SA, Sevene EJ, Hastings IM, Nosten F, McGready R. Antimalarial 23. Ngaimisi E, Habtewold A, Minzi O, Makonnen E, Mugusi S, Amogne W, drugs and pregnancy: safety, pharmacokinetics, and pharmacovigilance. et al. Importance of ethnicity, CYP2B6 and ABCB1 genotype for efavirenz Lancet Infect Dis. 2007;7:136–44. pharmacokinetics and treatment outcomes: a parallel-group prospec- 5. Anderson GD. Pregnancy-induced changes in pharmacokinetics: a tive cohort study in two sub-Saharan Africa populations. PLoS ONE. mechanistic-based approach. Clin Pharmacokinet. 2005;44:989–1008. 2013;8:e67946. 6. Isoherranen N, Thummel KE. Drug metabolism and transport during 24. Ngaimisi E, Mugusi S, Minzi O, Sasi P, Riedel K, Suda A, et al. Efect of pregnancy: how does drug disposition change during pregnancy and and CYP2B6 genotype on long-term efavirenz autoinduction what are the mechanisms that cause such changes? Drug Metab Dispos. and plasma exposure in HIV patients with or without tuberculosis. Clin 2013;41:256–62. Pharmacol Ther. 2011;90:406–13. 7. Jeong H. Altered drug metabolism during pregnancy: hormonal regula- 25. Piola P, Nabasumba C, Turyakira E, Dhorda M, Lindegardh N, Nyehangane tion of drug-metabolizing enzymes. Expert Opin Drug Metab Toxicol. D, et al. Efcacy and safety of artemether-lumefantrine compared with 2010;6:689–99. in pregnant women with uncomplicated Plasmodium falciparum 8. Staehli Hodel EM, Csajka C, Ariey F, Guidi M, Kabanywanyi AM, Duong malaria: an open-label, randomised, non-inferiority trial. Lancet Infect Dis. S, et al. Efect of single nucleotide polymorphisms in cytochrome P450 2010;10:762–9. isoenzyme and N-acetyltransferase 2 genes on the metabolism of arte- 26. Djimde A, Lefevre G. Understanding the pharmacokinetics of Coartem. misinin-based combination therapies in malaria patients from Cambodia Malar J. 2009;8(Suppl 1):S4. and Tanzania. Antimicrob Agents Chemother. 2013;57:950–8. 27. German PI, Aweeka FT. Clinical pharmacology of artemisinin-based 9. Mukonzo JK, Waako P, Ogwal-Okeng J, Gustafsson LL, Aklillu E. Genetic combination therapies. Clin Pharmacokinet. 2008;47:91–102. variations in ABCB1 and CYP3A5 as well as sex infuence quinine disposi- 28. Shokoples SE, Ndao M, Kowalewska-Grochowska K, Yanow SK. Mul- tion among Ugandans. Ther Drug Monit. 2010;32:346–52. tiplexed real-time PCR assay for discrimination of Plasmodium spe- 10. Piedade R, Gil JP. The pharmacogenetics of antimalaria artemisinin com- cies with improved sensitivity for mixed infections. J Clin Microbiol. bination therapy. Expert Opin Drug Metab Toxicol. 2011;7:1185–200. 2009;47:975–80. 11. Mirghani RA, Sayi J, Aklillu E, Allqvist A, Jande M, Wennerholm A, et al. 29. WHO. Methods and techniques for clinical trials on antimalarial CYP3A5 genotype has signifcant efect on quinine 3-hydroxylation in drug efcacy: genotyping to identify parasite populations. Geneva: Tanzanians, who have lower total CYP3A activity than a Swedish popula- World Health Organization. 2008. http://whqlibdoc.who.int/publica- tion. Pharmacogenet Genom. 2006;16:637–45. tions/2008/9789241596305_eng.pdf?ua 1. Accessed 30 Dec 2016. 12. Mugusi S, Ngaimisi E, Janabi M, Minzi O, Bakari M, Riedel KD, et al. Liver 30. Minzi OM, Ngaimisi E, Shewiyo DH, Sasi P,= Ignace AI. Interlaboratory enzyme abnormalities and associated risk factors in HIV patients on development and cross validation of a chromatographic method for Efavirenz-based HAART with or without tuberculosis co-infection in determination of lumefantrine in human plasma-a profcient capacity Tanzania. PLoS ONE. 2012;7:e40180. assessment of bioanalytical laboratories in East Africa. J Anal Bioanal Tech. 13. Gebeyehu E, Engidawork E, Bijnsdorp A, Aminy A, Diczfalusy U, Aklillu E. 2012;3:131. Sex and CYP3A5 genotype infuence total CYP3A activity: high CYP3A 31. WHO. Methods for surveillance of antimalarial drug efcacy. Geneva: activity and a unique distribution of CYP3A5 variant alleles in Ethiopians. World Health Organization. 2009. http://whqlibdoc.who.int/publica- Pharmacogenom J. 2011;11:130–7. tions/2009/9789241597531_eng.pdf. Accessed 1 Jan 2017. 14. Aklillu E, Djordjevic N, Carrillo JA, Makonnen E, Bertilsson L, Ingelman- 32. Tarning J, Kloprogge F, Dhorda M, Jullien V, Nosten F, White NJ, et al. Sundberg M. High CYP2A6 enzyme activity as measured by a cafeine Pharmacokinetic properties of artemether, dihydroartemisinin, lumefan- test and unique distribution of CYP2A6 variant alleles in Ethiopian popu- trine, and quinine in pregnant women with uncomplicated Plasmo- lation. OMICS. 2014;18:446–53. dium falciparum malaria in Uganda. Antimicrob Agents Chemother. 15. Habtewold A, Amogne W, Makonnen E, Yimer G, Nylen H, Riedel KD, et al. 2013;57:5096–103. Pharmacogenetic and pharmacokinetic aspects of CYP3A induction by 33. Ngaimisi E, Minzi O, Mugusi S, Sasi P, Riedel KD, Suda A, et al. Pharmacoki- efavirenz in HIV patients. Pharmacogenom J. 2013;13:484–9. netic and pharmacogenomic modelling of the CYP3A activity marker 16. Hatta FH, Lundblad M, Ramsjo M, Kang JH, Roh HK, Bertilsson L, et al. 4beta-hydroxycholesterol during efavirenz treatment and efavirenz/ Diferences in CYP2C9 genotype and enzyme activity between Swedes rifampicin co-treatment. J Antimicrob Chemother. 2014;69:3311–9. and Koreans of relevance for personalized medicine: role of ethnicity, 34. Diczfalusy U, Miura J, Roh HK, Mirghani RA, Sayi J, Larsson H, et al. 4Beta- genotype, smoking, age, and sex. OMICS. 2015;19:346–53. hydroxycholesterol is a new endogenous CYP3A marker: relationship to Mutagonda et al. Malar J (2017) 16:267 Page 10 of 10

CYP3A5 genotype, quinine 3-hydroxylation and sex in Koreans, Swedes 41. Lamba JK, Lin YS, Schuetz EG, Thummel KE. Genetic contribution to and Tanzanians. Pharmacogenet Genom. 2008;18:201–8. variable human CYP3A-mediated metabolism. Adv Drug Deliv Rev. 35. Williams JA, Ring BJ, Cantrell VE, Jones DR, Eckstein J, Ruterbories K, et al. 2002;54:1271–94. Comparative metabolic capabilities of CYP3A4, CYP3A5, and CYP3A7. 42. Wojnowski L, Hustert E, Klein K, Goldammer M, Haberl M, Kirchheiner Drug Metab Dispos. 2002;30:883–91. J, et al. Re: modifcation of clinical presentation of prostate tumors by a 36. Dawes M, Chowienczyk PJ. Drugs in pregnancy. Pharmacokinetics in novel genetic variant in CYP3A4. J Natl Cancer Inst. 2002;94:630–1. pregnancy. Best Pract Res Clin Obstet Gynaecol. 2001;15:819–26. 43. Dally H, Edler L, Jager B, Schmezer P, Spiegelhalder B, Dienemann H, et al. 37. Amirimani B, Ning B, Deitz AC, Weber BL, Kadlubar FF, Rebbeck TR. The CYP3A4*1B allele increases risk for small cell lung cancer: efect of Increased transcriptional activity of the CYP3A4*1B promoter variant. gender and smoking dose. Pharmacogenetics. 2003;13:607–18. Environ Mol Mutagen. 2003;42:299–305. 44. Aklillu E, Habtewold A, Ngaimisi E, Yimer G, Mugusi S, Amogne W, et al. 38. Rodriguez-Antona C, Sayi JG, Gustafsson LL, Bertilsson L, Ingelman- SLCO1B1 gene variations among Tanzanians, Ethiopians, and Europe- Sundberg M. Phenotype-genotype variability in the human CYP3A locus ans: relevance for African and worldwide precision medicine. OMICS. as assessed by the probe drug quinine and analyses of variant CYP3A4 2016;20:538–45. alleles. Biochem Biophys Res Commun. 2005;338:299–305. 45. Campbell MC, Tishkof SA. African genetic diversity: implications for 39. Lee JS, Cheong HS, Kim LH, Kim JO, Seo DW, Kim YH, et al. Screening of human demographic history, modern human origins, and complex genetic polymorphisms of CYP3A4 and CYP3A5 genes. Korean J Physiol disease mapping. Annu Rev Genom Hum Genet. 2008;9:403–33. Pharmacol. 2013;17:479–84. 40. Fukushima-Uesaka H, Saito Y, Watanabe H, Shiseki K, Saeki M, Nakamura T, et al. Haplotypes of CYP3A4 and their close linkage with CYP3A5 haplo- types in a Japanese population. Hum Mutat. 2004;23:100.

Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to find the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research

Submit your manuscript at www.biomedcentral.com/submit